Non-Parametric Mixture Model Based Evolution of Level Sets and Application to Medical Images
- 310 Downloads
We present a novel region-based curve evolution algorithm which has three primary contributions: (i) non-parametric estimation of probability distributions using the recently developed NP windows method; (ii) an inequality-constrained least squares method to model the image histogram with a mixture of nonparametric probability distributions; and (iii) accommodation of the partial volume effect, which is primarily due to low resolution images, and which often poses a significant challenge in medical image analysis (our primary application area). We first approximate the image intensity histogram as a mixture of non-parametric probability density functions (PDFs), justifying its use with respect to medical image analysis. The individual densities in the mixture are estimated using the recent NP windows PDF estimation method, which builds a continuous representation of discrete signals. A Bayesian framework is then formulated in which likelihood probabilities are given by the non-parametric PDFs and prior probabilities are calculated using an inequality constrained least squares method. The non-parametric PDFs are then learnt and the segmentation solution is spatially regularised using a level sets framework. The log ratio of the posterior probabilities is used to drive the level set evolution. As background to our approach, we recall related developments in level set methods. Results are presented for a set of synthetic and natural images as well as simulated and real medical images of various anatomical organs. Results on a range of images show the effectiveness of the proposed algorithm.
KeywordsNon-parametric probability density functions Finite mixture models Curve evolution Level sets
Unable to display preview. Download preview PDF.
- Bond, S., Joshi, N. B., Petroudi, S., & Brady, M. (2007). Estimating the mesorectal fascia in MRI. In: The proceedings of information processing in medical imaging—IPMI. Google Scholar
- Brainweb (1997). http://www.bic.mni.mcgill.ca/brainweb/.
- Ibanez, L., Schroeder, W., Ng, L., & Cates, J. (2005). The insight segmentation and registration toolkit software guide, version 2.4. Available at www.itk.org.
- ITK (2005). The insight segmentation and registration toolkit. Available at www.itk.org.
- Joshi, N. B. (2007). Non-parametric probability density function estimation for medical images. DPhil thesis, University of Oxford. Google Scholar
- Joshi, N. B., & Brady, M. (2005). A non-parametric mixture model for partial volume segmentation of MR images. In: The proceedings of British machine vision conference—BMVC. Google Scholar
- Joshi, N. B., & Brady, M. (2007). Non-parametric mixture model based evolution of level sets. In: The proceedings of international conference on computing: theory and applications—ICCTA (pp. 618–622). Google Scholar
- Joshi, N. B., & Brady, M. (2008). Non-parametric mixture model based evolution of level sets. In: Chanda B., Murthy C.A. (Eds.), Advances in intelligent information processing: tools and applications, World Scientific Publishing, Indian Statistical Institute, India, Platinum Jubilee Series. Google Scholar
- Kass, M., Witkin, A., & Terzopoulos, D. (1988). Snakes: active contour models. International Journal of Computer Vision, 321–331. Google Scholar
- Mumford, D., & Shah, J. (1990). Boundary detection by minimizing functionals, I*. In: Ullman S., Richards W. (Eds.), Image understanding (Chap. 2). Google Scholar
- Papoulis, A., & Pillai, S. U. (2002). Probability, random variables and stochastic processes. New York: McGraw-Hill. Google Scholar
- Paragios, N. (2000). Geodesic active regions and level set methods: contributions and applications in artificial vision. PhD thesis, University of Nice Sophia Antipolis/INRIA, France. Google Scholar
- Rathi, Y., Michailovich, O., Malcolm, J., & Tannenbaum, A. (2006). Seeing the unseen: segmenting with distributions. In: International conference on signal and image processing (SIP). Google Scholar
- Sijbers, J. (1998). Signal and noise estimation from magnetic resonance images. PhD thesis, University of Antwerp, Belgium. Google Scholar
- Zeydabadi, M., Zoroofi, R. A., & Soltanian-Zadeh, H. (2004). Multiresolution automatic segmentation of T1-weighted brain MR images. In: Proceedings of IEEE international symposium on biomedical imaging conference (pp. 165–168). Google Scholar